1. The Problems Known in the Art
Advances in biomaterials engineering have led to a number of highly visible and successful technologies in cardiovascular medicine and surgery, orthopedics, ophthalmology and dentistry. While the use of biomaterials has improved the quality of life for millions of people, there still remain problems with blood coagulation, fibrosis, infection, inflammatory reactions, degradation and rejection associated with currently available materials. Controlling biological interactions with materials is of great importance to the design of permanent as well as resorbable implants, cell and tissue scaffolds, and diagnostic probes, among others. The ability to control interactions between materials and bacteria is particularly important. Most materials used in medical devices are susceptible to bacterial adhesion. Once bacteria adhere to a solid surface, they generate and become embedded within a polysaccharide matrix to form a biofilm, within which, they are extremely difficult to combat. Both the host immune system and antimicrobials become less effective against the bacteria due to difficulty in penetrating the biofilm matrix and/or inactivation. In addition, bacteria in biofilms have more capacity to develop resistance to antimicrobials. To reduce morbidity and mortality due to device related infections, it is critical to prevent biofilm formation.
Catheters are routinely implanted in the bloodstream, urinary tract, chest, neck, abdomen, leg and spinal cord, and they are particularly susceptible to microbial infection, blood clotting, and occlusion that can begin within hours of implantation. An estimated ⅓ of catheters fail by infection and about ⅓ fail due to fibrin formation occluding the catheter. In either case, the most effective treatment currently is replacement with a similar device.
Implant-associated infections can result in systemic infections that in the worst-case scenario can lead to multiple organ failure and death, despite successful resolution of the original medical condition. The cost of such infections is high. For example, a single episode of central venous catheter-related bacteremia has been estimated to cost between $3,700 and $50,000 with an attributable mortality rate between 4 and 35%. Approximately 3 million central venous catheters are used each year in the United States and catheter related blood stream infections (CR-BSI) occur in over 200,000 patients with over 80,000 taking place in the intensive care unit (ICU). The cost of ICU infections alone is $296 million to $2.3 billion with between 2,400 and 20,000 deaths per year. Infection is also a major problem for dialysis patients. Over 450,000 people in the US alone have end stage kidney failure and require chronic hemodialysis. For these patients, vascular access procedures are a major cause of morbidity and mortality. AV grafts are used in about 42% of patients with an infection rate between 11 and 20%. The mortality rate due to infection of these grafts is between 12 and 22%.
Several catheter modification approaches have been evaluated for their ability to reduce the incidence of catheter-related blood stream infections (CR-BSI). The approaches can be divided into two categories. In one category, surfaces are modified to prevent bacterial adhesion. Many of these approaches involve minimization of adsorption and adhesion through steric repulsion and/or minimization of interfacial energy. In the case of catheters, adsorption of proteins, particularly fibrinogen, often leads to thrombus formation or development of a fibrin sheath and eventual occlusion. Several protein components of thrombus increase bacteria adhesion to catheters and there is an association with thrombus formation and CR-BSI. On the surfaces of hydrophobic materials, entropically driven, hydrophobic interaction dominates protein adsorption, and many research groups have shown that surfaces grafted with PEO are significantly less prone to adsorption and adhesion. Although hydrophilic coatings have been shown to reduce bacterial adhesion, problems with infection still occur.
In the second category, surfaces are modified or device materials are impregnated with agents that actively kill or prevent the growth of bacteria. Two commercially available short term catheters that fit into this category have been shown to reduce infection rates, one is chlorhexidine-silver sulfadiazine-impregnated (CSI) and the other is minocyline-rifampin impregnated (MRI). However these products pose a significant risk for developing drug resistant bacteria. This risk is lower for the antiseptic CSI catheters, but in vitro studies have found that exposure to chlorhexidine can result in increased bacteria resistance to it and other therapeutic antimicrobial agents. Furthermore, CSI is not effective if it needs to be in place for longer than three weeks, it is less effective than MRI, and serious anaphylactoid reactions associated with the use of these catheters have been reported in Japan. Although the antimicrobial catheters cost more than standard catheters, studies have demonstrated that there is an overall cost benefit of using them in high risk patients due to the high cost of treating CR-BSI. The added cost of antimicrobial catheters is between $25 and $34 per catheter, but the overall cost benefit is approximately $200 per catheter for CSI. Based on cost benefits and improved patient care, there is a strong motivation to use the antimicrobial catheters. However, these pose a very serious risk of furthering the development of resistant bacteria and are currently recommended for use only in high risk patients (in the ICU, on total parental nutrition, or immunosuppressed).
Unfortunately, these currently known technologies have not solved the needs in the art. Rather, what is needed is a material modification that (1) kills bacteria upon contact with the device such that antibacterial agents do not have to be released into the blood stream or surrounding tissue, (2) is biocompatible and will not adversely affect the patient when directly interfaced with the patient, (3) can be readily applied to a variety of materials and irregularly shaped objects, (4) prevents thrombus formation and occlusion, and (5) does not stimulate changes in bacteria that lead to resistance and therefore can be broadly used without compromising the effectiveness of clinical antibiotics.
2. Currently Known Block Copolymers
A different type of technology has been developed which uses PLURONIC® surfactants. PLURONIC® surfactants are generally high molecular weight polyoxyalkylene ether compounds that are water soluble. These block copolymer compounds have been used to immobilize bioactive entities at interfaces with good success. This approach utilizes a triblock copolymer (polyethylene oxide-polypropylene oxide-polyethylene oxide) that has been modified at the termini of the polyethylene oxide chains to allow for coupling to biomolecules. The modified copolymers are often referred to as end group activated polymers or “EGAP”. Various patents and patent applications have been filed related to these types of products including U.S. Pat. No. 6,087,452, U.S. Pat. No. 6,284,503, U.S. Pat. No. 6,670,199, U.S. Patent Application Publication No. 2004/0219541, U.S. Patent Application Publication No. 2004/0142011, U.S. Patent Application Publication No. 2005/0244456 and U.S. Patent Application Publication No. 2005/0106208 (which patents/applications are expressly incorporated herein by reference). Accordingly, for additional information regarding these products, the reader should consult these patents.
The EGAP molecules self-assemble on hydrophobic materials from aqueous solutions. The hydrophobic polypropylene oxide (“PPO”) center block forms a strong hydrophobic bond with the material while the polyethylene oxide (“PEO”) end blocks remain freely mobile in the fluid phase. Using this approach, a thick PEO brush-like layer is formed at the material surface that serves two important purposes. First, the PEO layer acts as a cushion between the peptide and the substrate preventing any denaturing of the peptide that might otherwise result from surface interaction and retaining peptide mobility. Second, the PEO layer prevents nonspecific adsorption of proteins or cells to which the surface might be exposed in subsequent use.
Previous studies have shown that the EGAP technology has advantages over other PEO based tethering technologies: it is very simple to apply to materials by a dip coating process, it can be applied to a variety of different types of materials and irregularly shaped objects, and it provides a mechanism to systematically vary protein surface concentration. Medical devices, implants, and tissue engineering scaffolds are prepared from a number of different material types due to the different mechanical, electrical, and optical properties that different applications require. Most materials that display optimal bulk properties for a given application do not have adequate biocompatibility. Direct chemical modification of materials to improve biocompatibility is complicated and can change material properties. In some cases, direct chemical modification is not feasible based on the material type or the irregular shape of an object. Direct adsorption of biomolecules on such surfaces often leads to denaturation of the biomolecules. The EGAP technology provides a simple and versatile solution to these challenges because it can be applied to a number of different materials by a simple dip coating process and can be used to immobilize proteins and peptides with retained activity.
3. Antimicrobial Peptides
Other research has been conducted regarding antimicrobial peptides that kill or slow the growth of microbes like bacteria, fungi, viruses, or parasites. Antimicrobial peptides that have been studied include defensins, cecropins, bacteriocins, and other natural or synthetic cationic peptides. Lantibiotics are one class of bacteriocins. They include the antimicrobial peptide nisin, and are antibiotic compounds that include one or more lanthionine rings. Over 40 lantibiotics are currently known, and more are being discovered each year. Analogs of antimicrobial peptides with improved activity or stability are also being developed [1]. The unique physical structure of lantibiotics, (e.g., double bonds, thioether rings, and unusual amino acid residues), makes these antimicrobial peptides highly reactive, and different in mode-of-action from traditional antibiotics. This suggests that they may remain effective despite the global increase of resistant bacterial strains. Lantibiotics also demonstrate wide variability in their inhibitory spectrums. Some lantibiotics, (e.g., nisin and subtilin) are active against Gram-positive bacteria, while other lantibiotics (e.g., cinnamycin) are active only against Gram-positive rods. Additionally, some lantibiotics have antiviral activity, (e.g., lanthiopeptin), some function as immunosuppressors, (e.g., mersacidin), and some (e.g., duramycin and ancovenin) can inhibit biomedically important enzymes.
The structure of nisin, by far the most extensively investigated lantibiotic with reference to biomaterials applications, is shown schematically in FIG. 1. Specifically, FIG. 1 shows that the N-terminal domain (residues 1-19) includes the three lanthionine rings labeled A, B, and C. FIG. 1 also shows, the C-terminal domain (residues 23-34) includes two lanthionine rings, identified as D and E. Additionally, FIG. 1 shows that a flexible hinge region (residues 20-22) connects the two domains. (In FIG. 1 Abu refers to 2-aminobutyric acid; Dha refers to dehydroalanine; Dhb refers to dehydrobutyrine; Ala-S-Ala refers to lanthionine; and Abu-S-Ala refers to β-methyllanthionine).
Nisin has had a long history as a potent and safe food preservative. It has been demonstrated that nisin can adsorb to synthetic surfaces, maintain activity, and kill cells that have adhered in vitro. While nisin has demonstrated activity against only Gram-positive bacteria, it can be an effective inhibitor of certain Gram-negative bacteria when used in combination with other compounds such as chelating agents. Staphylococcus aureus and Staphylococcus epidermidis are the most frequently encountered biomaterial-associated pathogens, and both are Gram-positive bacteria. Nisin has been shown to prevent microbial adhesion on endotracheal suction catheters in vitro (using Staphylococcus aureus, Staphylococcus epidermidis, and Enterococcus faecalis (Streptococcus faecalis) as indicator organisms), prompting further studies in vivo evaluating nisin-treated intravenous (IV) catheters in sheep and tracheotomy tubes in ponies. Catheters pretreated with nisin for long-term placement (7 days) did not retain antimicrobial activity, while short-term (3-5 h) IV catheters did. The exact duration of nisin activity on IV catheters remains unknown. There were no abnormalities on clinical examination of sheep during the experimental period, and no animal in either group developed catheter-related infection or venous thrombosis. Veins with short-term catheters also showed fewer and less severe histological abnormalities compared with controls, indicating a possible protective effect on vascular endothelium. As the first-ever preclinical trial of nisin-treated implantable materials, this study represented an important step toward development of protein antimicrobial films for implantable medical devices.
Lantibiotics have many characteristics that make them valuable for biomedical applications. Unlike typical peptides, lantibiotics contain dehydrated amino acid residues with electrophilic centers that readily react with nucleophilic groups, such as bacterial DNA and enzymes. The thioether rings found in all lantibiotics make the molecules more heat stable, much less affected by reducing agents, and less reactive toward free radicals than would disulfide bonds. Lantibiotics may offer a means for preventing the rise of resistant microorganisms, and since their mechanism of action is so dissimilar to that of traditional clinical antibiotics, cross-resistance is highly unlikely. There are some reports that repeated exposure to nisin can lead to changes in bacteria that can confer some weak resistance. However, the changes that occur also appear to make the bacteria weak compared to their nonresistant counterparts and more susceptible to antibiotics or the immune response. Nisin has been used broadly and extensively in food products for many years without problems arising due to development of resistance.
There are several different mechanisms through which lantibiotics can exert their antimicrobial effect. Type A lantibiotics (such as nisin) are linear molecules that are strongly cationic. They are highly surface active and kill susceptible bacteria through a multistep process that destabilizes the phospholipid bilayer of the cell and creates transient pores. The targeted bacterium is rapidly killed by efflux of ions and cytoplasmic solutes, such as amino acids and nucleotides, and subsequent dissipation of membrane potential. The depolarization of the cytoplasmic membrane results in an instant termination of all biosynthetic processes. Structural analyses have indicated that the hydrophilic groups of nisin interact with the phospholipids headgroups, and the hydrophobic side chains are immersed in the hydrophobic core of the membrane. The “wedge” model of pore formation takes such data into account but proposes that the peptides insert into the membrane without losing contact with the membrane surface, resulting in the formation of a short-lived (i.e., duration of milliseconds to seconds) pore. In the wedge model, pore formation is proposed to be caused by local perturbation of the lipid bilayer, whereby the hydrophobic residues of the peptide are inserted shallowly into the outer leaflet of the lipid bilayer. The “barrel-stave” model on the other hand proposes that nisin binds as a monomer and inserts into the lipid bilayer. The inserted monomers then aggregate laterally to form pores. Each of these models was proposed during a time when many questions remained concerning the involvement of cell surface factors, lifetime of the pore, and number of molecules required for pore formation. These have not all been answered, but recent research has revealed the importance of the cell wall precursor “lipid II,” and the functional importance of specific segments of the nisin molecule, in pore formation.